Sentiment Classification Research Articles

Personality-driven experience storage and retrieval for sentiment classification

Personality-driven experience storage and retrieval for sentiment classification

LexiSNTAGMM: an unsupervised framework for sentiment classification in data from distinct domains, synergistically integrating dictionary-based and machine learning approaches

LexiSNTAGMM: an unsupervised framework for sentiment classification in data from distinct domains, synergistically integrating dictionary-based and machine learning approaches

Microblog sentiment analysis method using BTCBMA model in Spark big data environment

Abstract Microblogs are currently one of the most well-liked social platforms in China, and sentiment analysis of microblog texts can help further analyze the realization of their media value; however, the current task of sentiment analysis based on microblog information suffers from low accuracy due to the large size and high redundancy of microblog data, a microblog sentiment analysis method using Bidirectional Encoder Representation from Transformers (BERT)–Text Convolutional Neural Network (TextCNN)–Bidirectional Gate Recurrent Unit (BiGRU)–Multihead-Attention model in Spark big data environment is proposed. First, the Chinese pre-trained language model BERT is used to convert the input data into dynamic character-level word vectors; then, TextCNN is used to effectively obtain local features such as keywords and pool the filtered features; then, BiGRU is introduced to quickly capture more comprehensive semantic information; finally, a multi-headed attention mechanism is implemented to emphasize the most significant features in order to accomplish the sentiment classification of microblog information task precisely. By comparing the existing advanced models, the proposed model demonstrates an improvement of at least 4.99% and 0.05 in accuracy and F1-score evaluation indexes, respectively. This enhancement significantly enhances the accuracy of microblog sentiment analysis tasks and aids pertinent authorities in comprehending the inclination of individual’s attitude toward hot topics. Furthermore, it facilitates a prompt prediction of topic trends, enabling them to guide public opinion accordingly.

Enhancing Social Media Insights: Leveraging Artificial Intelligence for Sentiment Analysis

Social media platforms have become indispensable channels for communication and interaction, offering a wealth of data that can provide valuable insights into public sentiments. However, analyzing sentiment on these platforms poses significant challenges due to the diverse and unstructured nature of user-generated content. Traditional Natural Language Processing (NLP) techniques struggle to accurately classify sentiments expressed through text, images, emoticons, and multimedia elements. Moreover, the informal and nuanced language used in Electronic Word of Mouth (eWOM) further complicates sentiment analysis. In response, this paper explores the role of Artificial Intelligence (AI) in improving sentiment analysis on social media. By leveraging Machine Learning (ML) algorithms trained on large datasets, AI can enhance the accuracy and efficiency of sentiment classification, providing decision-makers with actionable insights into the sentiment landscape of social media. Keywords: Social media, Sentiment analysis, Artificial Intelligence, Machine Learning, Natural Language Processing, Electronic Word of Mouth.

Media capture, captured: a new computational methodology to measure deteriorating media freedom

ABSTRACT The media are under increasing pressure by states to adopt a regime-friendly editorial line in many countries globally, raising the importance of understanding the conditions of successful and failed media capture. Unfortunately, the standard measurement of media capture – annual expert surveys – neither distinguishes between outlets nor offers the needed resolution to assess, e.g., the immediate impact of a new gag law. To remedy these shortcomings, we develop a computational measurement of media capture based on a comparison of nominally independent outlets and regime-owned outlets regarding media agenda and tone when referring to the regime. By relying on two unsupervised measurement methods – topic modelling and sentiment classification – the new method can assess the loss of editorial independence at the level of individual outlets on a monthly or even weekly basis. The methodology is validated by applying it to recent data from Nicaragua. There the two-pronged approach shows how outlets respond differently to regime pressure. While there still are limitations to consider, if further editorial aspects of media capture such as issue framing are added to the methodology, it has the potential of transforming the future study of media capture.

Transdisciplinary fine-grained citation content analysis: A multi-task learning perspective for citation aspect and sentiment classification

The diffusion of citation knowledge is an important measure of transdisciplinary scientific impact and the diversity of transdisciplinary citation content (sentences). Moreover, combining citation sentiment (CS) and citation aspect (CA) can help researchers identify the attitudes, ideas, or positions reflected in the evolution of scientific elements (e.g., theories, techniques, and methods). This is because of their use by scholars from different disciplines, paving the way toward transdisciplinary penetration and the development of domain knowledge through the proliferation of cited knowledge. However, most studies mainly address citation aspect classification (CAC) and citation sentiment classification (CSC) separately, ignoring their shared features of interactions. In this study, we construct a dataset for transdisciplinary citation content analysis using citations and academic full texts from the Chinese Social Sciences Citation Index (CSSCI), which includes 14,832 manually-annotated citations. Thereafter, we utilized the developed dataset to conduct a transdisciplinary fine-grained citation content analysis by combining CAC and CSC. The objective of the CAC task was to classify transdisciplinary citations into theoretical concepts (TC), methodological techniques (MT), and data information (DI), whereas the CSC task classified citations into positive, negative, and neutral classes. Furthermore, we leveraged a multi-task learning (MTL) model to perform CAC and CSC jointly and then compared its performance to those of several widely-used deep learning models. Our model achieved 83.10 % accuracy for CAC and 80.46 % accuracy for CSC, demonstrating its superiority to single-task systems. This indicates the strong correlation between the CAC and CSC of transdisciplinary citation tasks, benefiting from each other when learned concurrently. This new method can be used as an auxiliary decision support system to extend the analysis of transdisciplinary citation content.

A novel self-supervised sentiment classification approach using semantic labeling based on contextual embeddings

AbstractSentiment Analysis (SA) is a domain or context-oriented task since the sentiment words convey different sentiments in various domains. As a result, the domain-independent lexicons cannot correctly recognize the sentiment of domain-dependent words. To address this problem, this paper proposes a novel self-supervised SA method based on semantic similarity, contextual embedding, and Deep Learning Techniques. It introduces a new Pseudo-label generator that estimates the pseudo-labels of samples using semantic similarity between the samples and their sentiment words. It proposes two new concepts to calculate semantic similarity: The Soft-Cosine Similarity of a sample with its Positive words (SCSP) and the Soft-Cosine Similarity of a document with its Negative words (SCSN). Then, the Pseudo-label generator uses these concepts and the number of sentiment words to estimate the label of each sample. Later on, a novel method is proposed to find the samples with highly accurate pseudo-labels. Finally, a hybrid classifier, composed of a Convolutional Neural Network (CNN) and a Gated Recurrent Unit (GRU), is trained using these highly accurate pseudo-labeled data to predict the label of unseen data. The comparison of the proposed method with the lexicons and other similar existing methods demonstrates that the proposed method outperforms them in terms of accuracy, precision, recall, and F1 score.

Sentiment Analysis on Twitter through Machine Learning: A Comprehensive Approach with User-Centric Visualisations

Sentiment analysis is a crucial field that deals with the intricate task of identifying and systematically categorizing the various perspectives and opinions expressed within the original text. In today's digital age, social media platforms serve as a prolific source of data, inundated with a relentless stream of status updates, tweets, and content imbued with sentiments. Analysing the sentiments conveyed by users in this vast reservoir of data holds a pivotal role in comprehending the collective sentiments of the user community, dissecting dialogues, and aggregating viewpoints. This, in turn, can be instrumental in shaping strategies for commerce, conducting insightful political research, and gauging the pulse of communal activities. Examining sentiments on Twitter presents an increased difficulty because of the frequency of spelling errors, casual language, icons, and emojis. This research focuses on Twitter sentiment analysis, with a specific emphasis on a particular user account. The approach involves a combination of Python programming and Machine Learning techniques. By embarking on a comprehensive sentiment analysis journey within a specific domain, the aim is to discern the profound impact of that domain's data on sentiment categorization. Furthermore, this paper introduces a novel feature that enhances the organization of a user's most recent tweets and their presentation through visual aids such as graphs, charts, and word clouds. This visualization approach empowers a more intuitive and insightful exploration of the sentiments and trends embedded within the user's Twitter activity, facilitating a deeper understanding of their thoughts and emotions as expressed through their digital interactions.

Similar modality completion-based multimodal sentiment analysis under uncertain missing modalities

Recently, uncertain missing modalities in multimodal sentiment analysis (MSA) brings a new challenge for sentiment analysis. However, existing research cannot accurately complete the missing modalities, and fail to explore the advantages of the text modality in MSA. For the above problems, this work develops a Similar Modality Completion based-MSA model under uncertain missing modalities (termed as SMCMSA). Firstly, we construct the full modalities samples database (FMSD) by screening out the full modality samples from the whole multimodal dataset, and then predicting and marking the sentiment labels of each modality of the samples with three pre-trained unimodal sentiment analysis model (PTUSA). Next, for completing the uncertain missing modalities, we propose a set of missing modalities completion strategies based on the similar modalities selected from FMSD. For the completed multimodal data, we first encode the text, video and audio modality using the encoder of transformer, then we fuse the representation of text into the representations of video and audio under the guidance of a pre-trained model, thereby improving the quality of video and audio. Finally, we conduct sentiment classification based on the representations of text, video and audio with the softmax function respectively, and get the final decision with the decision-level fusion method. Based on benchmark datasets CMU-MOSI and IEMOCAP, extensive experiments have been conducted to verify that our proposed model SMCMSA has better performance than that of the state-of-the-art baseline models.The codes of our model are available at https://github.com/Astro2Sun/SMCMSA.

Leveraging machine learning and dimensionality reduction for sports and exercise sentiment analysis

The realm of sports and physical exercise is abuzz with a plethora of opinions, ranging from enthusiastic support to disheartening criticism. Delving into this sea of sentiment can be quite challenging due to the vastness and complexity of the data. This study proposes a novel approach that combines the power of dimensionality reduction with conventional machine learning techniques to effectively analyze sentiment in sports and exercise-related texts. We have encountered that the emotions regarding sports and exercise have been flooded because of the lockdown conditions and machines are not trained to analyze these emotions. The objective of this analysis is to enhance the accuracy of sentiment classification, enabling machines to better understand the nuances of human emotions expressed in sports and exercise contexts. Our experiments demonstrate that the proposed Dimensionality Reduction-Enhanced Machine Learning approach outperforms traditional methods in identifying the sentiment in sports and exercise-related texts. This technique achieves a significant improvement in weighted F1-score of 5 %, indicating a more precise and reliable sentiment classification. Among the conventional machine learning algorithms employed, Naïve Bayes, Support Vector Machines (SVM), Logistic Regression & Artificial neural network emerged as the most effective methods, achieving a weighted F1-score of respectively 0.92,0.925, 0.935, and 0.91. This study paves the way for a deeper understanding of public opinion during pandemic times and their effect on sports and physical exercise, enabling organizations and individuals to make informed decisions regarding marketing strategies, program development, and community engagement. The ability to accurately discern sentiment from a vast array of sports and exercise-related texts holds immense potential for fostering a more engaged and inclusive sporting landscape.

Filter channel network based on contextual position weight for aspect-based sentiment classification

Filter channel network based on contextual position weight for aspect-based sentiment classification

Opinion Extraction using Hybrid Learning Algorithm with Feature Set Optimization Approach

Evolution in engineering and technology added large size of data storage and transmission through the web application over the internet. This huge amount of data primarily used for exchange of information in between users and devices and in secondary aspects it has utilization as feedback, ratings and reviews that is supporting in generation of useful information of products, services, incidents etc. The data as opinion, feedback, view & suggestion is explored, organized & analyzed for selection of appropriate options. Sentiment analysis using the opinion extraction is a challenging task that is based on feature extraction and the concepts of Natural Language Processing that is applied in identification of the opinions of a user in terms of positive, neutral or negative ratings hidden in the form of comments typed as the text. Presently many data-processing based feature evaluation techniques for opinion extraction are used for solving the issues faced under sentiment classification applications. This article is based on development and application of algorithms for opinion extraction from text data available on web resources by K-Nearest Neighbor (KNN), Support vector machine (SVM) and hybrid of both named as SVM+KNN for classification of multi-label opinions from extracted text from review data of Twitter and Amazon. The performance of all the classification models (KNN, SVM and SVM+KNN) on both datasets is evaluated in terms of different parameters.

Evaluation of adjective and adverb types for effective Twitter sentiment classification.

Twitter, the largest microblogging platform, has reported more than 330 million active users in recent years. Many users express their sentiments about politics, sports, products, personalities, etc. Sentiment analysis has emerged as a specialized branch of machine learning in which tweets are binary-classified to provide sentimental insights. A major step in sentiment classification is feature selection, which primarily revolves around parts of speech (POS). Few techniques merely focused on single features such as adjectives, adverbs, and verbs, while other techniques examined types of these features, such as comparative adjectives, superlative adjectives, or general adverbs. Furthermore, POS as linguistic entities have also been studied and extensively classified by researchers, such as CLAWS-C7. For sentiment analysis, none of the studies conceptualized all possible POS features under similar conditions to draw firm conclusion. This research is centered on the following objectives: 1) examining the impact of various types of adjectives and adverbs that have not been previously explored for sentiment classification; 2) analyzing potential combinations of adjectives and adverbs types 3) conducting a comparison with a benchmark dataset for better classification accuracy. To assess the concept, a renowned human annotated dataset of tweets is investigated. Results showed that classification accuracy for adjectives is improved up to 83% based on the general superlative adjective whereas for adverbs, comparative general adverb also depicted significant accuracy improvement. Their combination with general adjectives and general adverbs also played a substantial role. The unexplored potential of adjectives and adverb types proved better in accuracy against state-of-the-art probabilistic model. In comparison to lexicon-based model, proposed research model overruled the dependency of lexicon-based dictionary where each term first needs to be matched for semantic orientation. The evident outcomes also help in time reduction aspect where huge volume of data need to be processed swiftly. This noteworthy contribution brought up significant knowledge and direction for domain experts. In the future, the proposed technique will be explored for other types of textual data across different domains.

A Comparative Analysis of Feature Selection Algorithms in Cross Domain Sentiment Classification

Background: Cross-domain Sentiment Classification is a well-researched field in sentiment analysis. The biggest challenge in CDSC arises from the differences in domains and features, which cause a decrease in model performance when applying source domain features to predict sentiment in the target domain. To address this challenge, several feature selection methods can be employed to identify the most relevant features for training and testing in CDSC. Method: The primary objective of this study is to perform a comparative analysis of different feature selection methods on the various CDSC tasks. In this study, statistical test-based feature selection methods using 18 classifiers for the CDSC task has been implemented. The impact of these feature selection methods on Amazon product reviews, specifically those in the DVD, Electronics, Kitchen, and TV domains, has been compared. Total 12x18 experiments were conducted for each feature selection method by varying source and target domain pairs from the Amazon product reviews dataset and by using 18 classifiers. Performance evaluation measures are accuracy and f-score. Results: From the experiments, it has been inferred that the CSDC task depends on various factors for a good performance, from the right domain selection to the right feature selection method. We have concluded that the best training dataset is Electronics as it gives more precise results while testing in either domain selected for our study. Conclusion: Cross-domain sentiment analysis is a dynamic and interdisciplinary field that offers valuable insights for understanding how sentiment varies across different domains.

An analysis of customer perception using lexicon-based sentiment analysis of Arabic Texts framework

Sentiment Analysis (SA) employing Natural Language Processing (NLP) is pivotal in determining the positivity and negativity of customer feedback. Although significant research in SA is focused on English texts, there is a growing demand for SA in other widely spoken languages, such as Arabic. This is predominantly due to the global reach of social media which enables users to express opinions on products in any language and, in turn, necessitates a thorough understanding of customers’ perceptions of new products based on social media conversations. However, the current research studies demonstrate inadequacies in furnishing text analysis for comprehending the perceptions of Arabic customers towards coffee and coffee products. Therefore, this study proposes a comprehensive Lexicon-based Sentiment Analysis on Arabic Texts (LSAnArTe) framework applied to social media data, to understand customer perceptions of coffee, a widely consumed product in the Arabic-speaking world. The LSAnArTe Framework incorporates the existing AraSenTi dictionary, an Arabic database of sentiment scores for Arabic words, and lemmatizes unknown words using the Qalasadi open platform. It classifies each word as positive, negative or neutral before conducting sentence-level sentiment classification. Data collected from X (formerly known as Twitter, resulted in a cleaned dataset of 10,769 tweets, is used to validate the proposed framework, which is then compared with Amazon Comprehend. The dataset was annotated manually to ensure maximum accuracy and reliability in validating the proposed LSAnArTe Framework. The results revealed that the proposed LSAnArTe Framework, with an accuracy score of 93.79 %, outperformed the Amazon Comprehend tool, which had an accuracy of 51.90 %.

GREEN HYDROGEN: A SOCIAL MEDIA INTELLIGENCE ANALYSIS

Given the current state of climate change and the fight against it, green hydrogen has the potential to be the energy vector of the future within a context of decarbonisation. In this way, fossil fuels might be replaced by green hydrogen in those sectors that are most difficult to decarbonise. Therefore, in recent years it is becoming more and more common to hear about green hydrogen in society and the great advantages of using it, although, it also has some disadvantages, mostly related to higher costs and higher energy consumption. With this in mind, the purpose of this paper is to use social media intelligence for the topical issue of green hydrogen; i.e. the objective is to gather and analyse the social perception of it. For this purpose, taking the social media platform Twitter as our source of information, we extracted 625,794 green hydrogen tweets for the years 2020, 2021 and 2022. On the one hand, a social network analysis was carried out, obtaining different metrics and identifying the main communities that have been talking about green hydrogen in the digital sphere over the last three years. On the other hand, using artificial neural networks, a sentiment classification model was applied to the tweets with the aim of detecting the emotion generated by society. Accordingly, results were obtained and interpreted through a novel methodological combination. The network analysis revealed that the conversation taking place is not polarized and is a reflection of the digital agora that brings together citizens, media actors and actors from the economic-political space. In addition, the sentiment analysis conducted shows that the overall digital discussion is positive. Keywords: Green hydrogen, social media, text mining, artificial neural networks, social network analysis, social perception, technology and social change, Twitter

Enhancing aspect-based sentiment analysis with BERT-driven context generation and quality filtering

Fine-grained sentiment analysis, commonly referred to as aspect-based sentiment analysis (ABSA), has garnered substantial attention in both academic and industrial circles. ABSA focuses on unveiling the sentiment orientation associated with specific entities or attributes within textual data, resulting in a more precise depiction of intricate emotional nuances. However, due to the extensive range of applications for ABSA, certain domains face challenges such as constrained dataset sizes and the absence of exhaustive, high-quality corpora, leading to issues like few-shot learning and resource scarcity scenarios. To address the issue of limited training dataset sizes, one viable approach involves the utilization of text-based context generation to expand the dataset. In this study, we amalgamate Bert-based text generation with text filtering algorithms to formulate our model. Our model fully leverages contextual information using the Bert model, with a particular emphasis on the interrelationships between sentences. This approach effectively integrates the relationships between sentences and labels, resulting in the creation of an initial data augmentation corpus. Subsequently, filtering algorithms have been devised to enhance the quality of the initial augmentation corpus by eliminating low-quality generated data, ultimately yielding the final text-enhanced dataset. Experimental findings on the Semeval-2014 Laptop and Restaurant datasets demonstrate that the enhanced dataset enhances text quality and markedly boosts the performance of models for aspect-level sentiment classification.

Evaluasi Aplikasi Raileo Melalui Analisis Sentimen Ulasan Playstore Dengan Metode Naive Bayes

Abstrak The Raileo application is a staffing platform owned by PT. KAI, functions as a personnel data management system. Effective application development requires data as a basis, and one source of data that can be utilized is user reviews. User reviews provide valuable information regarding application performance, user needs, and security aspects. However, challenges arise in managing review data which often contains sarcasm, creating ambiguous meaning and lowering accuracy levels. This research proposes a solution by applying sentiment analysis using Naive Bayes logarithms to 1047 Raileo review data. This method produces an accuracy rate of 94%, with positive and negative sentiment classification. The research results show the words that appear most frequently in Raileo reviews, such as "eror", "sulit", "titik presensi", "titik absen", "titik lokasi", "bug", "lemot," "gagal", "mantap", "bagus", "oke", "mudah", "mempermudah", "mantul", "lengkap","keren","ok", "inovatif", "inovasi", "semoga", "sukses", dan "membantu". These words can be used as a key to analyze all the sentiments contained in the review. In addition, this research identifies "presence point" as the highest negative sentiment word that needs attention in further development. From this sentiment analysis research, the Raileo application produces the highest sentiment value, namely positive sentiment

Analyse the Enhancement of Sentiment Analysis in Arabic by doing a Comparative Study of Several Machine Learning Techniques

Abstract: Sentiment analysis is a crucial component of natural language processing that seeks to determine the emotional sentiment expressed in a given text. This study investigates sentiment analysis in the Arabic language through a comprehensive approach that integrates traditional machine learning methods with sophisticated deep learning models. We examine the efficacy of conventional algorithms such as Support Vector Machines (SVM) and Naive Bayes, as well as sophisticated neural network architectures such as Convolutional Neural Networks (CNNs), Long Short-Term Memory networks (LSTMs), and the Arabic variant of Bidirectional Encoder Representations from Transformers (BERT). The primary novelty of our approach is in the ensemble method, which combines many approaches to enhance the precision of sentiment categorization in Arabic text. To address the particular challenges presented by Arabic sentiment analysis, such as the intricate structure of the language and the diverse regional variations, we utilize a tailored preprocessing pipeline to effectively handle the nuances of Arabic text. Our comprehensive analysis of various datasets demonstrates that the ensemble technique outperforms individual model benchmarks and offers novel insights into the interplay between different machine learning paradigms in Arabic NLP. The results emphasize the ability of hybrid approaches to improve Arabic sentiment analysis, providing a solid basis for future research and practical applications in understanding the sentiments of Arabic consumers. This study is a significant addition to the expanding domain of Arabic Natural Language Processing (NLP). This resource offers a comprehensive and advanced methodology for utilizing machine learning and deep learning methods to comprehend and analyse the intricate aspects of sentiment in the Arabic language

Exploring the customer experience of Vietnam airlines via sentiment analysis

The sentiment analysis of social media has gained much attention due to the succinct nature of comments, enabling access to public opinions. Using this technique, the present study aimed to explore customer sentiments regarding their experiences with Vietnam Airlines, the national carrier. According to the Skytrax evaluation, Vietnam Airlines achieved a 4-star rating out of 5, indicating the need for service improvements to attain perfection. The significance of this research lies in providing a comprehensive perspective on customer experience evaluations. To achieve this, we utilized the Valence Aware Dictionary for Reasoner (VADER) to classify sentiments expressed in social networking forum data. The dataset comprised 670 reviews from the Skytrax forum, which underwent a data pre-processing phase. The results exhibited favorable accuracy in detecting ternary and multiple sentiment classes. The comparative analysis between visual customer ratings and their corresponding comments yielded similar outcomes. Additionally, the results highlighted the need for a re-evaluation of the in-flight entertainment system.